Teoria de Sistemas Cinzas aplicada ao estudo de variáveis micrometeorológicas em uma região de Cerrado no estado de Mato Grosso

• Ano de defesa: 2013
• Autor(a) principal: Rocha, Patrick Siqueira da
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Mato Grosso; Brasil; Instituto de Física (IF); UFMT CUC - Cuiabá; Programa de Pós-Graduação em Física Ambiental
• Palavras-chave em Português: Sistemas dinâmicos; Análise de incidência cinza; Preenchimento de falhas; CNPQ::CIENCIAS EXATAS E DA TERRA::FISICA; Dynamical systems; Grey incidence analysis; Gap filling
• Link de acesso: http://ri.ufmt.br/handle/1/3311

The development of mathematical tools that make the study of open and dynamical system possible is the base of Environmental Physics researches. Based on this statement, this works aim on the apply of Grey System Theory to analyse micrometeorological variables collected, using snsors installed in a tower, at an area of cerrado (brazilian savannah), near the city of Cuiabá. We developed a methodology for the application of Gray System Theory, which consists in the creation of computational routines implementing differents parts of this theory. One of these routines is used to gap filling, that occurs due to problems during the collection phase. This process of gap filling presented results with average accuracy higher than 95%. We also developed routines to pre-processing these date, preparing it to meet theory requirements. After this, grey incidence degree was calculated to 14 variables, pairwise. We used these indexes to see if the theory could identify phenomena that occurs in these systems. As a final step, it was possible to stablish, through these indexes and the phase space construction, that only two variables carry enough information of the system in the month of August 2012 (dry-season). The same does not hold true for December 2012 (wet-season) where the coupling of moisture disturbs the system, creating the need of inclusion for more variables.